Particle Tomography Results (Demo)

This repository contains all scripts that were used to create the reconstructions and plots shown in the paper Particle Tomography: A Grid-Free Approach to Tomographic Reconstruction. The actual Particle Tomography algorithm is implemented here. (This link will become active once the paper is publicly available.)

To reproduce the results shown in the paper, follow these steps:

1. Clone the repository

git clone https://github.com/microscopic-image-analysis/particle_tomography_results.git
cd particle_tomography_results

2. Set up the environment

We provide a conda environment specification to ensure all dependencies are installed correctly.

conda env create -f environment.yml
conda activate particle-tomography-results

3. Install Particle Tomography

Follow the instructions given here to install the Particle Tomography package.

4. Run the scripts

From the repository root, run:

python scripts/main.py

This will run all benchmarks on the protein dataset and generate plots. Internally, it executes

benchmark.run_all_benchmarks()
plot.plot_protein(show_3d_volumes=False)

5. Notes

• We only provide the protein dataset in the data folder. The vesicle and thinfilm dataset from “Accurate real space iterative reconstruction (RESIRE) algorithm for tomography” can be downloaded from here, and the platinum dataset from "Nanomaterial datasets to advance tomography in scanning transmission electron microscopy" from here. In benchmark.py the parts corresponding to the other datasets have been commented out.
• Resire reconstructions were obtained with the scripts in scripts/reconstruct/resire_matlab_scripts. To obtain plots for these, the reconstructed volumes must be placed in out/RESIRE/name_of_dataset.